PEMETAAN KERAPATAN DAN PENERAPAN METODE DIFERENSIASI OBIA UNTUK DIFERENSIASI JENIS MANGROVE DI KAWASAN TANJUNG PEMANCINGAN, KOTABARU

Abstract

This study aims to identify mangrove species and map the density of mangrove vegetation in the Tanjung Pemancingan area, Kotabaru, using an object-based classification method (OBIA) applied to Sentinel-2 imagery and Unmanned Aerial Vehicle (UAV) data. Mangroves play a crucial role in protecting coastlines from erosion and serving as habitats for various species, making an in-depth understanding of mangrove distribution and types essential for coastal conservation and environmental management. The OBIA method allows for more accurate mapping by considering texture, shape, and more complex spatial patterns compared to traditional pixel-based methods. This study employs the Support Vector Machine (SVM) algorithm in the classification process to enhance the accuracy of mangrove species identification. The analysis utilizes Sentinel-2 satellite imagery with a spatial resolution of 10x10 meters and UAV data for higher resolution. The results show that the NDVI values for mangroves in the study area range from -0.30 to 0.686, which were classified into three canopy density classes: sparse (-0.30 to 0.026), moderate (0.027 to 0.356), and dense (0.357 to 0.686). The OBIA method combined with the SVM algorithm successfully discriminated between seven mangrove species with an overall accuracy (OA) of 72.46%. The identified mangrove species include Avicennia alba, Avicennia marina, Avicennia officinalis, Avicennia rumphiana, Bruguiera gymnorhiza, Rhizophora apiculata, and Sonneratia alba, with Avicennia rumphiana being the most dominant species, covering an area of 13.87 hectares. The mangrove vegetation density was successfully mapped, providing valuable information that can be used in conservation planning, coastal resource management, and ecotourism development in the area. Furthermore, these results have significant implications for further research in mangrove ecosystem monitoring and the application of remote sensing technology in environmental management.